Can supervised exercise prevent treatment toxicity in patients with prostate cancer initiating androgen-deprivation therapy: a randomised controlled trial Prue Cormie*, Daniel A. Galvão*, Nigel Spry*†‡§, David Joseph*†‡, Raphael Chee‡§, Dennis R. Taaffe*¶, Suzanne K. Chambers*,**††‡‡ and Robert U. Newton* *Edith Cowan University Health and Wellness Institute, Edith Cowan University, §Genesis Cancer Care, Joondalup, † Department of Radiation Oncology, Sir Charles Gairdner Hospital, ‡Faculty of Medicine, University of Western Australia, Nedlands, ¶School of Environmental and Life Sciences, University of Newcastle, Ourimbah, **Griffith Health Institute, Griffith University, Southport, ††Cancer Council Queensland, Brisbane, and ‡‡Prostate Cancer Foundation of Australia, Sydney, Australia

Objective To determine if supervised exercise minimises treatment toxicity in patients with prostate cancer initiating androgen-deprivation therapy (ADT). This is the first study to date that has investigated the potential role of exercise in preventing ADT toxicity rather than recovering from established toxicities.

Patients and Methods Sixty-three men scheduled to receive ADT were randomly assigned to a 3-month supervised exercise programme involving aerobic and resistance exercise sessions commenced within 10 days of their first ADT injection (32 men) or usual care (31 men). The primary outcome was body composition (lean and fat mass). Other study outcomes included bone mineral density, physical function, blood biomarkers of chronic disease risk and bone turnover, general and prostate cancer-specific quality of life, fatigue and psychological distress. Outcomes were compared between groups using analysis of covariance adjusted for baseline values.

Results

fat with group differences of −1.4 kg (P = 0.001), −0.9 kg (P = 0.008) and −1.3% (P < 0.001), respectively. Significant between-group differences were also seen favouring the exercise group for cardiovascular fitness (peak oxygen consumption 1.1 mL/kg/min, P = 0.004), muscular strength (4.0–25.9 kg, P ≤ 0.026), lower body function (–1.1 s, P < 0.001), total cholesterol: high-density lipoprotein-cholesterol ratio (–0.52, P = 0.028), sexual function (15.2, P = 0.028), fatigue (3.1, P = 0.042), psychological distress (–2.2, P = 0.045), social functioning (3.8, P = 0.015) and mental health (3.6–3.8, P ≤ 0.022). There were no significant group differences for any other outcomes.

Conclusion Commencing a supervised exercise programme involving aerobic and resistance exercise when initiating ADT significantly reduced treatment toxicity, while improving social functioning and mental health. Concurrent prescription of supervised exercise when initiating ADT is therefore advised to minimise morbidity associated with severe hypogonadism.

Compared to usual care, a 3-month exercise programme preserved appendicular lean mass (P = 0.019) and prevented gains in whole body fat mass, trunk fat mass and percentage

Keywords

Introduction

including abdominal adiposity) [2–5], alterations in insulin, lipids and C-reactive protein (CRP) [5,6], as well as reductions in bone mineral density (BMD) [2,7] have been reported, contributing to an increased risk of cardiovascular, metabolic and bone diseases [8–10]. Furthermore, men undergoing ADT experience significant declines in sexual health [11,12], reduced functional capacity [13,14] and increased

Androgen-deprivation therapy (ADT) has established therapeutic benefit for men with prostate cancer; however, ADT is associated with considerable adverse effects widely recognised throughout the literature [1]. Negative changes in body composition (decreased lean mass, increased fat mass

BJU Int 2015; 115: 256–266 wileyonlinelibrary.com

prostate cancer, androgen deprivation, exercise, resistance, aerobic

© 2014 The Authors BJU International © 2014 BJU International | doi:10.1111/bju.12646 Published by John Wiley & Sons Ltd. www.bjui.org

Exercise programme to reduce treatment toxicity when initiating ADT

psychological distress [15], culminating in significant declines in health-related quality of life (HRQL) [12,14]. Effective management strategies for ADT toxicities are clearly required to minimise treatment-related morbidity. Clinical trials have established the efficacy of appropriate exercise in ameliorating many of the adverse effects of ADT [16–20], leading to exercise being recognised as a management strategy for men undergoing ADT [1,21]. Exercise programmes combining resistance and aerobic components performed at moderate–high intensity have resulted in improvements in body composition, chronic disease risk factors, functional capacity, sexual health, fatigue and HRQL [16–20]. However, this evidence is limited to men on long-term ADT with an average treatment time of ≈14 months [16–20] and above [22–26]. The development of toxicities have been reported to be most pronounced during the initial months of ADT [3,4,7,14], raising the question of whether initiating exercise at the commencement of ADT can immediately attenuate or even prevent treatment toxicities. Our previous work has confirmed that men on ADT respond to exercise similarly regardless of treatment duration with the notable exception of changes in adiposity [27]. Compared with patients with chronic ADT exposure (average ≈25 months), patients within 6 months of commencing ADT (average ≈3 months) significantly increased body fat mass and triglyceride levels over a 12-week exercise intervention [27], highlighting the particularly strong drive to increased adiposity during the initial months of ADT [3]. Given the prevalence of long-term morbidity [8–10], the potential to mitigate adverse effects at the onset of ADT has a high degree of clinical relevance, as well as clear implications for patient HRQL. Therefore, the purpose of this trial was to determine if supervised exercise reduces treatment toxicity in patients with prostate cancer initiating ADT. To date, all investigations into the role of exercise in counteracting adverse effects of ADT have been of rehabilitative intent. This is the first trial which has investigated the potential role of exercise in preventing ADT toxicity rather than recovering from established toxicities.

Patients and Methods One hundred and twenty-six men aged 46–80 years scheduled to commence ADT for the treatment of prostate cancer were referred by oncologists and urologists in Perth, Western Australia from June 2011 through to October 2012, and screened for participation in the study (Fig. 1). Participants were not involved with any other clinical trial or exercise trial. Participants had a histological diagnosis of prostate cancer, were beginning treatment with leuprorelin acetate depot (Lucrin®), were anticipated to remain on ADT for at least the next 3 months and obtained written medical clearance from their physician (GP). Patients were excluded if they had

previously received ADT, had established bone metastatic disease, were unable to walk 400 m unassisted or had musculoskeletal, cardiovascular and/or neurological disorders that could inhibit them from exercising (as determined by the patient’s physician). This protocol was approved by the University Human Research Ethics Committee and all participants provided written informed consent. Experimental Design A two-armed prospective randomised controlled trial design was implemented. After familiarisation and baseline testing sessions, participants were randomised into the two arms: exercise (32 men) or usual care (31 men). Stratification for age (≤ or >70 years) was applied and participants were randomised in an allocation ratio of 1:1 using a random assignment computer program. The project coordinator and exercise physiologists involved in assigning participants to groups were ‘blinded’ to the allocation sequence. Participants randomised to the usual-care group received no intervention but were offered the exercise programme after the completion of the intervention period. All participants maintained standard medical care for the treatment of prostate cancer and were instructed to maintain their customary activity and dietary patterns throughout the intervention period. Exercise Intervention The exercise intervention involved twice weekly exercise sessions for 3 months in one of six exercise clinics across Perth and regional Western Australia (Bunbury). Sessions were conducted in small groups of 8–10 participants supervised by accredited exercise physiologists. The sessions were ≈60 min in duration and involved moderate–high intensity aerobic and resistance exercises, as well as standard warm-up and cool-down periods. The aerobic exercise component included 20–30 min of cardiovascular exercise using various modes, e.g. walking or jogging on a treadmill, cycling or rowing on a stationary ergometer or exercising on a cross trainer machine. Target intensity was set at approximately 70–85% of estimated maximum heart rate. The resistance exercise component involved eight exercises that targeted the major upper and lower body muscle groups (leg press, leg extension, leg curl, calf raise, chest press, lat pulldown, biceps curl and triceps extension). Intensity was manipulated from 6–12 repetition maximum (RM; i.e. the maximal weight that can be lifted 6–12 times, which is equivalent to ≈60–85% of one repetition maximum [1RM]) using 1–4 sets per exercise. To ensure the progressive nature of the programme, participants were encouraged to work past the specific RM prescribed and the resistance was increased by a 5–10% increment for the next set/training session if the participant exceeded the target RM. Prescription of both aerobic and resistance exercise components was progressive and modified according to individual response. Session rating of perceived exertion © 2014 The Authors BJU International © 2014 BJU International

257

Cormie et al.

Directly referred by oncologists and assessed for eligibility (n = 126)

Fig. 1 CONSORT diagram.

Excluded (n = 63) • Already received ADT injection (n = 26) • Could not complete 400-m walk (n = 3) • GP decline (n = 1) • Declined to participate (n = 10) • Too far to travel (n = 6) • Other reasons (n = 17)

Randomised (n = 63)

Allocated to exercise (n = 32)

Allocated to usual care (n = 31)

Discontinued intervention (n = 1) • ADT side-effects (nauseous, dizzy and fatigued; n = 1)

Lost to follow-up (n = 7) • Wanted to start exercising (n = 4) • Too far to travel (n = 2) • Time constraints (n = 1)

Analysed (n = 32) • Missing baseline data for DXA (n = 4), pQCT (n = 4), SOT (n = 4) and blood markers (n = 3)

Analysed (n = 31) • Missing baseline data for DXA (n = 1), pQCT (n = 1), and SOT (n = 1)

(RPE) was recorded immediately after the completion of each exercise session to monitor the perceived intensity of the exercise [28]. Participants were encouraged to supplement the supervised exercise sessions with home-based aerobic exercise with the aim of accumulating a total of at least 150 min of moderate intensity aerobic exercise each week. The exercise intervention was designed in accordance with international guidelines [29].

BMD Areal BMD (g/cm2) of the hip (femoral neck), lumbar spine (L2–4) and whole body were assessed by DXA using standard protocols [2]. Volumetric BMD (mg/cm3) of the left tibia was assessed by peripheral quantitative CT (pQCT; XCT3000, Stratec, Pforzheim, Germany). Standard scanning and analysis procedures were used to determine total BMD at 4% of tibial length [30].

Outcome Measures

Physical function

Study endpoints were assessed at baseline and after the intervention (3 months). Participants completed a familiarisation session 3 days before baseline testing involving all physical function assessments.

A series of standard tests were used to assess physical function. Cardiovascular fitness was assessed using the 400-m walk test to estimate peak oxygen consumption (VO2peak) [31]. Maximal strength of the lower and upper body was determined using the 1RM in the leg press, chest press and seated row exercises [16]. Lower body function was assessed using the repeated chair rise and stair climb tests [16]. Usual and fast pace 6-m walks evaluated ambulatory ability while the 6-m backwards walk was used to assess dynamic balance [16]. Static balance was determined using the sensory organisation test (SOT) performed on a Neurocom Smart Balancemaster

Body composition Regional and whole body lean mass and fat mass were derived from whole body dual-energy X-ray absorptiometry scans (DXA; Hologic Discovery A, Waltham, MA, USA). Appendicular lean mass, trunk adiposity and estimated visceral fat mass were assessed using standard procedures [2].

258

© 2014 The Authors BJU International © 2014 BJU International

Exercise programme to reduce treatment toxicity when initiating ADT

(Neurocom, OR, USA) [16]. Falls self-efficacy was also determined using the Activities-Specific Balance Confidence scale (ABC score; a higher score represents greater balance confidence) [32]. Blood pressure and blood biomarkers A validated oscillometric device (HEM-705CP, Omron Corporation, Japan) was used to record brachial blood pressure in triplicate after resting in a supine position for 5 min. Fasting blood samples were obtained on the day of, or 1–3 days after the initial ADT injection. CRP, cholesterol, triglycerides, insulin, glucose, glycated haemoglobin, markers of bone formation and resorption (alkaline phosphatase, procollagen type 1 N-terminal propeptide [P1NP], N-telopeptide, N-telopeptide/creatinine ratio), vitamin D as well as testosterone and PSA levels were assessed. Samples were obtained and analysed commercially by an accredited Australian National Association of Testing Authorities laboratory (Pathwest Diagnostics, WA, Australia). Patient-reported outcomes (PROs) A series of questionnaires with sound psychometric properties were used to assess general and prostate cancer-specific HRQL, fatigue and psychological distress. The Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36) was used to assess general HRQL status (higher scores represent greater HRQL) [33]. The European Organisation for Research and Treatment of Cancer prostate cancer-specific module (QLQ-PR25) evaluated disease-specific HRQL across domains relating to urinary, bowel and treatment-related symptoms (lower scores represent less symptom severity), as well as sexual activity and sexual function (higher scores represent greater function) [34]. Fatigue was assessed using the Functional Assessment of Chronic Illness Therapy-Fatigue (FACIT-Fatigue) questionnaire (higher scores indicate less fatigue) [35]. The Brief Symptom Inventory-18 (BSI-18) was used to assess psychological distress across the domains of depression, anxiety, somatisation and global distress severity (lower scores represent less distress) [36]. Statistical Analyses Sample size calculations were based on change in body composition as indicated by lean mass. Previous research in men receiving ADT indicated a change in lean mass of ≈0.8 kg (standard deviation of ≈1 kg) after a 3-month exercise intervention [16]. A priori, 25 participants per group were required to achieve 80% power at an α level of 0.05 and to show a difference in lean mass between the groups after intervention. Based on previous experience we anticipated an attrition rate of up to 20%. Therefore, to adequately ensure that we had sufficient participant numbers at the end of the intervention, 63 participants were

randomised to the study arms (exercise group 32 men; usual-care group 31 men). Data were analysed using SPSS Statistics 21 (IBM Australia Ltd, NSW, Australia). Analyses including standard descriptive statistics, independent and paired t-tests, chi-square, analysis of covariance (ANCOVA) adjusted for baseline values. An intention-to-treat approach was used for all analyses using maximum likelihood imputation of missing values (expectation maximisation). Estimates for imputed values were based on the baseline value and change over time according to group allocation for all outcomes, as well as age for all physical function outcomes. All tests were two-tailed with statistical significance set at an α level of 0.05.

Results Patient Characteristics The two groups were well balanced with no significant differences in characteristics at baseline (Table 1). Furthermore, there were no significant differences between groups at baseline in any of the outcome measures assessed (Tables 2 and 3). One participant from the exercise group withdrew from the intervention due to feeling too nauseous, dizzy and fatigued to attend the exercise sessions. Seven participants in the usual-care group were lost to follow-up due Table 1 Baseline characteristics of participants. Baseline variables

Age, years Height, cm Weight, kg BMI, kg/m2 Number of comorbidities* Number of medications Self-rated health† Tertiary education Married Past smoker Current smoker Meets physical activity guidelines‡ Gleason score Time since diagnosis, months Time since ADT injection, days LHRH agonist + antiandrogen Radiation during the intervention Duration of radiation, weeks Previous radiation Time since radiation, years Previous prostatectomy Time since prostatectomy, years Previous chemotherapy§

Exercise (n = 32)

Usual care (n = 31)

P

69.6 ± 6.5 172.2 ± 6.1 86.7 ± 13.7 29.3 ± 4.5 1.6 ± 1.3 3.5 ± 2.0 2.7 ± 0.8 5 (15) 24 (72) 15 (45) 1 (3) 9 (27) 7.3 ± 0.8 16.6 ± 27.0 6.2 ± 1.6 15 (45) 7 (22) 5.0 ± 4.8 2 (6) 1.8 ± 1.3 4 (12) 3.1 ± 3.2 1 (3)

67.1 ± 7.5 172.5 ± 6.6 88.3 ± 10.5 29.6 ± 2.6 1.5 ± 1.3 3.1 ± 2.2 2.5 ± 0.8 8 (26) 24 (77) 14 (45) 1 (3) 9 (29) 7.7 ± 1.2 10.4 ± 14.4 5.6 ± 2.0 19 (61) 8 (26) 8.9 ± 7.1 1 (3) 1.0 ± 0 9 (29) 4.7 ± 6.5 0 (0)

0.150 0.853 0.609 0.698 0.707 0.469 0.364 0.290 0.665 0.987 0.964 0.876 0.229 0.290 0.235 0.367 0.714 0.319 0.573 0.335 0.093 0.647 0.321

Results presented as mean ± SD or number of participants (percentage of participants). *Cardiovascular disease, hypertension, diabetes, osteoporosis and dyslipidaemia. † Self-rated health – 1, excellent; 2, very good; 3, good; 4, fair; 5, poor. ‡Self-reported physical activity level of ≥150 min/week. §Treatment for chronic lymphocytic leukaemia ~5 years before study enrolment.

© 2014 The Authors BJU International © 2014 BJU International

259

Cormie et al.

Table 2 Body composition, BMD and physical function values and the group difference in change over the intervention. Measure

Mean (SD) value at baseline Exercise group

Body composition, DXA Whole body total mass, kg Whole body lean mass, kg Appendicular lean mass, kg Whole body fat mass, kg Whole body percentage fat, % Trunk fat mass, kg Visceral fat mass, g BMD (DXA/pQCT) Hip BMD, g/cm2 Spine BMD, g/cm2 Whole Body BMD, g/cm2 Tibia BMD, mg/cm3 Physical function VO2peak, mL/kg/min 400-m walk, s Leg press 1RM, kg Chest press 1RM, kg Seated rRow 1RM, kg Repeated chair rise, s Stair climb, s 6-m walk – usual pace, s 6-m walk – fast pace, s Balance – 6-m backwards walk, s Balance – SOT, score Balance confidence, ABC score

Mean (SD) value at 3 months

Usual-care group

Exercise group

Usual-care group

86.2 (14.4) 56.6 (6.8) 23.7 (3.0) 26.9 (8.9) 30.6 (5.2) 14.8 (5.1) 912.6 (259.6)

88.4 (10.6) 58.7 (6.6) 24.9 (3.1) 26.9 (6.0) 30.3 (4.1) 15.2 (4.0) 926.4 (252.2)

85.0 (11.7) 56.0 (6.0) 23.5 (2.8) 26.3 (7.1) 30.5 (4.4) 14.3 (4.2) 874.0 (247.9)

87.8 (9.9) 57.3 (6.3) 24.3 (2.9) 27.8 (5.5) 31.4 (3.9) 15.5 (3.8) 922.2 (240.8)

0.861 (0.112) 1.206 (0.182) 1.202 (0.128) 322.1 (46.2)

0.866 (0.096) 1.196 (0.167) 1.186 (0.094) 329.5 (48.6)

0.849 (0.100) 1.190 (0.168) 1.193 (0.110) 318.3 (42.5)

0.855 (0.089) 1.191 (0.163) 1.182 (0.093) 321.4 (42.4)

22.1 (3.5) 260.9 (44.3) 134.3 (50.0) 44.8 (12.8) 71.9 (13.3) 11.6 (2.4) 5.27 (1.86) 4.36 (0.65) 3.25 (0.49) 15.42 (7.60) 76.2 (4.9) 90.3 (16.8)

23.2 (3.4) 248.5 (36.8) 143.6 (52.4) 50.4 (16.2) 78.0 (16.4) 11.2 (2.3) 4.60 (1.20) 4.04 (0.57) 3.14 (0.47) 13.95 (5.14) 77.7 (5.0) 93.6 (10.6)

22.7 (3.8) 254.4 (42.8) 157.9 (52.9) 47.2 (13.1) 74.8 (13.8) 10.5 (2.0) 5.08 (1.65) 4.32 (0.62) 3.17 (0.45) 15.10 (7.40) 76.1 (6.3) 90.3 (16.8)

22.7 (3.6) 253.0 (40.6) 141.7 (9.6) 47.4 (15.4) 75.7 (12.7) 11.1 (2.2) 4.87 (1.12) 4.20 (0.38) 3.15 (0.43) 14.27 (5.50) 78.6 (5.4) 93.3 (6.3)

Adjusted group differences in mean change over 3 months* Mean (95% CI)

−0.8 (−2.3 to 0.7) 0.7 (−0.1 to 1.6) 0.4 (0.1 to 0.7) −1.4 (−2.3 to −0.6) −1.3 (−1.7 to −0.8) −0.9 (−1.5 to −0.2) −35.9 (−80.8 to 9.0) 0.000 (−0.025 to 0.024) −0.009 (−0.029 to 0.012) −0.002 (−0.013 to 0.009) 4.0 (−2.5 to 10.5) 1.1 (0.4 to 1.9) −10.3 (−18.0 to −2.7) 25.9 (15.0 to 36.8) 4.8 (1.6 to 7.9) 4.0 (0.5 to 7.6) −1.1 (−1.5 to −0.6) −0.30 (−0.67 to 0.07) −0.09 (−0.26 to 0.08) −0.06 (−0.17 to 0.05) −0.47 (−2.05 to 1.10) −0.9 (−2.9 to 1.1) −3.2 (−9.6 to 3.3)

P

0.290 0.078 0.019 0.001